Roboflow + iOS
Run your Roboflow model on iPhone and iPad with the Roboflow Mobile iOS SDK, for on-device object detection and instance segmentation in native apps.
How to connect iOS to Roboflow
5 stepsTrain a model in Roboflow
Train an object detection or instance segmentation model and note its model ID and version number.
Add the SDK with CocoaPods
Run pod init in your Xcode project, add pod 'Roboflow' to the Podfile, run pod install, then open the .xcworkspace file.
Initialize with your API key
Create a RoboflowMobile instance with your Roboflow API key.
Load the model
Call load with your model ID and version. The SDK offers completion-handler and async variants.
Run detection on images
Pass an image to the model's detect method and draw the returned predictions in your app. The roboflow-ios-starter app shows camera setup and drawing.
What teams build with iOS
Detect objects in a live camera app
Run detection on device so results keep up with the camera feed.
Build an augmented reality feature
Use on-device predictions to place content on objects the camera sees.
Count items from a photo
Ship an app that counts what it sees, like the CashCounter example that counts US coins and bills.
Questions
Object detection, and instance segmentation on iOS 18 or later. Classification and semantic segmentation are not supported for SDK deployment.
iOS 15.4 at minimum, or iOS 18.0 for instance segmentation. Devices older than the iPhone 8 fall back to the GPU, which uses more energy.
Yes. Roboflow publishes RoboflowExpoExample, a React Native Expo integration example, alongside the native starter app.
Our docs document CocoaPods installation only.
Why are companies choosing Roboflow?
The end-to-end platform for vision AI: create datasets, train models and deploy to production.
Flexible deployment modes
Move from a scalable cloud API to edge hardware like NVIDIA Jetson without managing complex dependencies.
Integrated workflow builder
Combine multiple AI models, custom business logic and third-party integrations into one production-ready pipeline in a low-code editor.
Model monitoring and alerts
Dashboards track confidence, inference time and request volume for your deployed models, and email alerts tell your team when confidence suddenly drops.
